Closed zmstone closed 1 month ago
if we implement a in-place downsample of the current table, there is no need for a new table schema. although not so efficient, but after downsampled, the number of data points will no longer be big enough to have a visible performance impact.
if we implement a in-place downsample of the current table, there is no need for a new table schema. although not so efficient, but after downsampled, the number of data points will no longer be big enough to have a visible performance impact.